AI · 2h ago
Tokenless routes AI agent calls to cheaper models automatically
Tokenless, a YC-backed startup, launches an API gateway that dynamically routes AI agent requests between models to cut costs. The system matches Claude Fable 5 performance at half the price by querying multiple models simultaneously and switching based on progress. Founders claim the cache-aware routing technique is novel, with plans to add more models like Kimi K3.
Meridian48 take
The cost-saving pitch is strong, but the real test is whether the router's performance holds up under diverse, real-world agent workloads beyond the founders' benchmarks.
Read the full reporting
Launch HN: Tokenless (YC S26) – Automatic model switching to save money →
Hacker News
ai-cost-optimizationmodel-routing